Free AI Chatbot: 6 Apps Compared, Paid From $8 [2026]

A free AI chatbot in October 2026 still comes with a trade. You get a capable assistant for $0, and the vendor keeps the strongest model, the longest sessions and the newest agent features for paying customers. ChatGPT’s pricing page shows the pattern in a plain feature matrix: GPT-6.1 Sol is marked “No” on Free and on Go, and “Yes” on Plus.

The same split shows up across the six assistants most readers weigh first, which are ChatGPT, Claude, Gemini, Grok, Meta AI and Microsoft Copilot. This comparison checks each one against the vendors’ own pages as of October 24, 2026. It ranks them by job instead of by one blended score, because the free plans differ far more in what they hold back than in how well they chat.

One caveat before the details. None of the official pages we reviewed publishes a hard daily message count for its free plan. We already measured free-tier message caps in a separate test, so this piece skips the cap-counting and asks a different question. Which free chatbot should you open first for your task, and what does the first paid step cost when you hit a wall?

What a Free AI Chatbot Means in October 2026

“Free” now covers three different deals. The first is a plan that costs $0 and stays that way, with a lighter model and a quota the vendor never states in numbers. The second is a free feature inside an app you already use, which is how Meta AI and Copilot work. The third is a free tier that pays for itself through ads or data, and that category is growing. Reports from this week say OpenAI expanded ChatGPT advertising to 26 additional countries, though we could not verify which countries or which plans the change touches.

Every vendor in this comparison refreshed its model line in the last two months, so last season’s advice is stale. OpenAI announced GPT-6 Sol and GPT-6 Luna in October, and its deployment page says they replace GPT-5.6 Sol and GPT-5.6 Luna inside ChatGPT. Anthropic shipped Claude Opus 5.5 on September 22, Claude Sonnet 5.5 on September 28 and Claude Haiku 5.5 on October 7, according to its release notes. Google’s launch post for Gemini 4 Argon is dated October 13. The xAI documentation dates Grok 4.7 to September 21. Meta announced Muse Wonder and Muse Wonder Full on October 12.

We scored six assistants and left two out. Perplexity and DeepSeek both belong in any broad conversation about free chatbots, but our fact check this week could not confirm their current plan names or limits from official pages. We do not print numbers we cannot source. If you lean on either, read the vendor’s own plan page before you trust anything in a roundup, this one included.

The three prices of free

Free tiers cost something even at $0. The first price is capability. You get the lighter model, and the vendor decides when you have used enough. The second is attention, since ads are spreading and free accounts are the natural place to show them. The third is data. Each vendor sets its own rules for whether free chats can help train future models, and those rules change with little notice, so the settings screen matters more than any article.

A last definitional point. A free AI chatbot is not the same thing as the best free AI chatbot for you. The best one depends on whether you write, code, search, study or create images, and on which ecosystem you already live in. The rest of this article follows that idea, with a spec table first, then one section per assistant, then data, pricing, scenarios, recommendations and a migration guide.

Specs Table: Six Free AI Chatbots Side by Side

The table below uses only details that appeared on the vendors’ own pages or in at least two independent checks. Where a vendor does not publish a number, the cell says so. That gap is itself a finding, because it tells you the free quota can change without notice.

Three patterns jump out. First, every vendor reserves its newest flagship for paid users, even though the free plan runs a current-generation model. Second, the entry price for a paid upgrade clusters tightly around $8 for the consumer-first products (Grok Plus at $8 and Meta One Core at $7.99). Third, only Grok publishes a full ladder of consumer prices in one place, which makes it the easiest to budget for.

Read the model-family row with care. A vendor naming a new model does not mean free users can pick it. In almost every case the free plan gets the lighter sibling of the new family, and the heavy model arrives in the first paid tier.

ChatGPT Free: GPT-6 Luna for Everyone, GPT-6.1 Sol for Plus

ChatGPT remains the default answer to the question of which free chatbot to try. OpenAI’s pricing page lists six plan names: Free, Go, Plus, Pro, Business and Enterprise. Free and Go users can reach GPT-6 Luna in the desktop app, per the same page. GPT-6.1 Sol, the stronger model in the matrix, carries a “No” for both Free and Go and a “Yes” for Plus.

The Luna and Sol names deserve a short history. OpenAI announced GPT-6 Sol and GPT-6 Luna in October as the replacements for GPT-5.6 Sol and GPT-5.6 Luna in ChatGPT. Reports also say OpenAI retired GPT-5.5 from ChatGPT, ChatGPT Work and Codex on October 14, although API access stayed. If a guide you read still calls GPT-5 or GPT-5.6 the newest ChatGPT model, it is out of date. Our flagship head-to-head covers how the top-end GPT-6 models stack up against rivals, so here we stay on the free side of the fence.

The Pro line has changed too. OpenAI’s help center names three Pro tiers, Pro 100, Pro 200 and Pro 500. Pro 200 stays at $200 a month, and the help article says eligible existing subscribers keep their previous usage allowance for a limited transition period, after which the included allowance changes. The Pro 100 and Pro 500 prices appear in only one source we checked, so treat $100 and $500 as reported figures. Pro 500 reportedly includes “Ultrafast” access.

For a free user, the practical question is how long Luna keeps up with your task. It should handle everyday work such as drafts, summaries, explainers and quick code questions. It will run out of road on long reasoning chains, large file jobs and anything that needs a Sol-class or Pro-class model. In practice, free plans across the industry wall off uploads, image generation, voice and research features first, and ChatGPT’s matrix of “No” entries for the stronger models fits that pattern. When you start losing time to those walls, Plus is the step to price.

Two newer changes affect free users directly. OpenAI reportedly added Locked Chats on October 23, which protects selected conversations behind a PIN or device biometrics and covers Free, Go, Plus, Pro, Business, Enterprise and Edu plans. And the advertising expansion mentioned above means a free ChatGPT account may show sponsored content in more countries than it did a month ago. Check your region’s current behavior before you assume a clean interface.

Claude Free: Strong Writing Models, Opaque Limits

Anthropic’s consumer product is simply Claude, available through Claude.ai on the web, iOS and Android. A free plan exists. Anthropic’s official pages in our check do not confirm the free plan’s current message limits, model access list or reset schedule, so any article that prints a precise free Claude message count today is guessing.

What we can confirm is the model line. Claude Opus 5.5 arrived in September as Anthropic’s flagship for agentic coding and knowledge work. Claude Sonnet 5.5 followed on September 28, and Anthropic says it runs more than 30% faster than Sonnet 5 and costs up to 30% less on typical workloads. Claude Haiku 5.5 arrived on October 7 as the company’s fastest small model, and Anthropic’s own site calls it the cheapest, fastest and most capable small model it has released. The Claude release notes list all three.

On pricing, the Claude pricing page shows Pro at $20 when billed monthly, with a discount for annual billing. Max and Team plans sit above that. The page also lists API credits as a perk of some higher plans, which matters if you build things on the side. Anthropic announced a monthly API credit for Max and Team subscribers on October 7.

The feature that shows the free-versus-paid line most clearly is Claude Duet. Anthropic opened a research preview of Duet on October 21, an app in which Claude works as a proactive agent, and it is limited to Pro and Max users. Free users cannot try it. The new class of always-on agents across Claude, OpenAI’s Dots and Meta’s Muse is a separate buying decision from picking a free chatbot, but it explains why paid tiers keep pulling ahead.

Writers and analysts tend to like Claude because its responses read like edited prose rather than bullet lists. That is a subjective judgment. The practical advice is to test it on your own drafts. Paste a real paragraph, ask for a tighter version, and see whether you accept the edit. If you hit a usage wall mid-session, save your work and come back, or price Pro.

Pew’s survey found 9% of US teens used Claude in 2025, the lowest share of the five assistants it tracked. That says little about quality and a lot about distribution, since Claude has no built-in placement inside a social app or a phone operating system.

Gemini Free: The Frontier Model Stays Behind a Paywall

Google’s consumer product remains the Gemini app, and Gemini also powers AI Mode in Google Search. The headline model is Gemini 4 Argon. Google’s launch coverage describes Argon as built for hard problems, and the main launch post carries an October 13 date. Google aims it at long-running work such as software engineering, research and cybersecurity.

For API users, Google set an introductory price of $2 per million input tokens and $10 per million output tokens, with cached input discounted by 95%. After November 30, 2026, the listed price becomes $4 per million input tokens and $20 per million output tokens. That doubling is on Google’s own launch page, and it matters if you plan to build on Argon. For consumers, Google’s rollout coverage says Argon reached the Gemini app and AI Mode for Google AI Ultra and Google AI Pro subscribers, and Ultra subscribers also get it in Google Antigravity.

So what does the free Gemini plan run? Google’s plan descriptions mention Gemini 3.6 Flash and Gemini 3.1 Pro, but the pages we reached do not say which model a free account defaults to, so we leave that cell unconfirmed. What matters for your decision is the direction of travel. Google follows the same pattern as OpenAI and Anthropic, with the frontier model on paid plans and a lighter one for free accounts.

Where Gemini wins on a free plan is integration. If you live inside Gmail, Docs, Drive and Search, the free assistant meets you where you already work, and AI Mode in Search gives you an answer engine at no cost. Google also reportedly began rolling out business-oriented Gemini features, including a business landing page and marketing and finance skills, to Gemini app users globally on October 21.

Pew’s 2025 survey put Gemini at 23% of US teens, second only to ChatGPT among the five assistants it named. That reach fits Google’s strategy of making the free assistant a default part of Android and Search, and it is a good reason to test Gemini alongside ChatGPT before you commit to either.

Grok Free: The Only Vendor With a Full Public Price Ladder

Grok is the cleanest case for budgeting, because xAI’s subscription documentation lays out every consumer tier with a price. xAI introduced The X Subscription on October 12, which folds Grok, Cursor and X Premium benefits into one ladder. Here is what the pages list:

• Free, $0 a month: limited daily Grok access, a starter allowance for Cursor coding agents, and standard X features.
• Plus, $8 a month or $80 a year: more Grok and Cursor usage, an X checkmark, creator tools and fewer ads.
• Premium, $30 a month: described as including SuperGrok, Cursor Pro and X Premium+, which xAI previously listed at $90 a month combined.
• Super, $100 a month: the Premium benefits with higher usage.
• Ultra, $200 a month: the highest listed allowance.

The documentation describes usage in relative terms rather than message counts. One source on our sheet puts Premium at 4 times the Plus allowance, Super at 15 times and Ultra at 50 times, but it appears in a single place, so treat those multiples as reported. The free tier stays limited rather than unlimited.

The model behind the app is Grok 4.7, which xAI’s documentation lists with a 500,000-token context window, text and image input, configurable reasoning and a knowledge cutoff of May 2026. API pricing is $2 per million input tokens and $6 per million output tokens for prompts under 200,000 tokens, and $4 and $12 at or above that size. The company also launched Grok Voice 3.0, priced at $0.08 per minute through the speech-to-speech API, unchanged from version 2.0. You can see the current free offer on the Grok site.

A branding note matters here. The documentation and news pages now carry the name SpaceXSI, and an October 22 post announced a “SpaceX Super Intelligence Partner Network.” The sources do not say whether that is a rename of xAI, a merger step or a partner program, so we keep calling the maker xAI and flag the ambiguity. Grok Build, the app-building tool, reportedly became available on every plan including Free.

Grok’s free tier suits curious users who already spend time on X and want to try an assistant bundled with it. It suits budget upgraders too, since $8 is one of the two cheapest verified steps into higher usage, alongside Meta One Core at $7.99.

Meta AI and Copilot: Free Because You Are Already There

These two chatbots follow a different logic from the first four. Neither Meta nor Microsoft needs you to open a dedicated chat app. Each wants its assistant inside the products you already use.

Meta stated on September 15 that the core Meta AI experience stays free, and it launched Meta One the same day as a subscription service layered on top. Meta One Core costs $7.99 a month and adds more Restyle, voice effects, creative tools and Meta AI image and video generation. Meta One Premium costs $19.99 a month and offers the highest individual level of access to Meta AI creation features. Single-product plans such as Instagram Plus and WhatsApp Plus start at $2.99 a month, and creator and business bundles start at $14.99. Meta’s help pages describe the AI allowance as variable and subject to testing, and one help page reportedly notes that Meta One was in limited testing and varied by location.

The model story is moving. On October 12 Meta announced Muse Wonder and Muse Wonder Full, which it calls its most capable models to date. Muse Wonder Full is available to Muse Code subscribers, and Meta said Muse Wonder would soon power Muse, initially for subscribers. Reach is Meta’s real advantage. Its Q3 2026 results report that its family of apps averaged 3.64 billion daily active people in September 2026, and Pew’s survey had Meta AI at 20% of US teens. For a direct matchup against Microsoft and a European rival, our look at Copilot, Meta AI and Mistral’s Le Chat covers the details.

Copilot is harder to pin down. Our sources gave no verified current details on its consumer plans, free-tier limits or pricing, so we list no figures for it here and suggest checking Microsoft’s own pages before you rely on it.

Both assistants make sense as free options if you already sit in their ecosystems. Instagram and WhatsApp users get Meta AI without an extra download, and Windows and Microsoft 365 users get Copilot in the tools they open every day. Neither is the first pick for someone who wants a stand-alone chat habit.

Usage Data and Price Benchmarks: What the Numbers Say

No standard benchmark covers free tiers. Academic leaderboards test flagship models, and free plans run lighter siblings with unpublished limits. So we lean on three kinds of evidence that do exist: survey data on who uses what, official API price lists as a cost benchmark, and vendor distribution figures.

Start with adoption. Pew Research Center’s December 2025 report on teens and AI chatbots states:

“In 2025, 59% of U.S. teens reported using ChatGPT, compared with 23% for Gemini, 20% for Meta AI, 14% for Copilot, and 9% for Claude.”

That is from Pew Research Center, and it is a 2025 snapshot of teens, not today’s adult market. The same Pew report adds a second data point:

“In 2025, 64% of U.S. teens reported ever using an AI chatbot, including approximately 30% who reported using chatbots daily.”

Both figures predate the October 2026 model refreshes, but they show where free usage concentrates. For newer share data, see our usage-share comparison of the four biggest assistants.

The second benchmark is price per million tokens. API list prices are the cleanest public measure of what a vendor charges for intelligence, and they hint at what a free tier costs the vendor to serve.

OpenAI’s October announcement reportedly lists GPT-6 Sol at $2 input and $10 output and GPT-6 Luna at $0.10 input and $0.50 output. Anthropic reportedly lists Claude Haiku 5.5 at the same $0.10 and $0.50 for prompts up to 100,000 tokens. Both are single-source figures, so we keep them out of the table. The overlap is still telling. When two vendors price their small models identically, it is a fair inference that small models carry most free-tier traffic. For how caching changes these bills, see our API cache pricing comparison.

The third benchmark is distribution. Meta’s Q3 release shows 3.64 billion daily active people across its apps in September 2026, and free Meta AI reaches a large share of them without a download. That distribution gap explains why a mid-pack model can still win on reach.

Since no lab benchmarks the free tiers, the most reliable test is your own seven-day log. Pick five tasks you do every week. Run each through two assistants. Note the time, how many edits you made, and the moment any assistant stops you. By day seven you will have a better data set than any review, because it measures your work on your plan.

Pricing Ladder: From $0 to the Top Tier

Paid upgrades follow one rule across the industry. The first paid step costs about as much as a streaming service. The top tier costs far more: Grok’s Ultra tier, at $200 a month, is 25 times its Plus tier at $8. The table lists every price we could verify on a vendor page and flags what we could not.

Two takeaways from the table. First, if your goal is the cheapest verified upgrade with a real chatbot attached, Grok Plus at $8 and Meta One Core at $7.99 lead, though Meta’s tier mostly buys creative generation rather than a smarter model. Second, Claude’s annual option trims the Pro price by roughly 17%. We calculated that from the two prices on Anthropic’s page ($20 monthly against $200 for a year).

Regional prices differ. Meta’s French announcement reportedly lists Core at 6.99 euros and Premium at 20.99 euros a month, and Meta warns that benefits, pricing and availability vary by region, app and account. Check the number your own app store shows before you budget.

We left some cells blank on purpose. OpenAI’s pricing page confirmed plan names but did not expose Go, Plus, Business or Enterprise prices in what we fetched, and the Google pages we reached did not state the Google AI Pro and Ultra prices. Open each vendor’s page before you buy, because tiers and trial offers are changing fast this autumn. Our broader subscription price comparison tracks the paid tiers in more depth if you are weighing a real upgrade.

A sensible rule of thumb: stay free until a concrete limit costs you more than the entry price in time. If a one-hour delay saves you nothing, do not pay. If you miss a deadline once a month, $8 to $20 is cheap insurance.

Seven Real-World Scenarios

These scenarios turn the plan facts above into concrete choices. They are illustrative walk-throughs, not case studies, and none of the details in them are test results.

Study and writing

Scenario 1, the student with a literature review. A student needs a 10-page review drafted from six papers. ChatGPT Free is the natural first stop, since 59% of the teens in Pew’s survey already use it and GPT-6 Luna should handle summaries well. The risk is the file-upload wall that free plans tend to impose. Splitting the papers across two sessions solves it, and Claude Free makes a good second opinion on the finished draft.

Scenario 2, the freelance writer on deadline. A writer polishing a long article benefits from Claude’s prose style. Free usage limits are the obstacle, and Anthropic does not publish them, so the writer works in sections, saves progress and keeps a backup assistant open in another tab. Claude Pro at $20 becomes worth it the first time a client deadline collides with a usage wall.

Scenario 3, the Gmail-and-Docs researcher. Someone who already works in Google tools should try Gemini first, because AI Mode in Search and the Gemini app sit one click from their documents. They should expect a lighter model on the free plan and keep Argon in mind for a paid month if the research gets deep.

Work and building

Scenario 4, the hobby developer. A developer building a small app wants free coding help. Grok Build reportedly works on every plan including Free, and the Free tier includes a starter allowance for Cursor coding agents. Paste the generated code into your editor, run your tests, and treat the output as a draft.

Scenario 5, the Microsoft 365 office worker. Copilot suits people who live in Word, Excel and Teams. Business buyers should note the December 1 changes. New Copilot Business licenses default to usage-based billing with a 4,000-credit monthly limit per user, which an administrator can adjust.

Creating and privacy

Scenario 6, the small shop owner making promo images. A shop owner posting to Instagram gets Meta AI free in the app and can step up to Meta One Core at $7.99 for more image and video generation. That price looks small next to most stand-alone image tools, though the allowance is variable and subject to testing.

Scenario 7, the privacy-conscious user. Someone who shares a phone with family wants private chats. Review the data and training settings in each app before you paste anything sensitive, and check whether the app lets you lock or hide past conversations.

Which Free Chatbot for Which Job: Seven Recommendations

Choose by task, not by brand loyalty. Our picks, with the reasoning behind each:

1. General everyday answers: ChatGPT Free. It runs GPT-6 Luna, has the broadest usage in Pew’s data (59% of US teens) and the widest plan ladder to grow into. Skip it if you object to ads.
2. Long-form writing and editing: Claude Free. The Claude 5.5 family is current, and the writing style suits drafting. Skip it if you need a published, predictable quota.
3. Google Workspace users: Gemini Free. Integration with Search and Docs beats any raw model edge. Keep Argon for a paid month if needed.
4. Hobby coding and tinkering: Grok Free. Grok Build and the Cursor starter allowance are reportedly included on the free plan, and Grok 4.7 has a 500,000-token window at the API level. Skip it if you want a mature ecosystem of plug-ins.
5. Social and creative content: Meta AI. Free inside Instagram, WhatsApp and Facebook, with Meta One Core at $7.99 for heavier generation. Skip it for serious analytical work.
6. Microsoft 365 workflows: Copilot. Best when your files already live in Microsoft’s cloud. Check licensing before you roll it out to a team.
7. Cheapest path to more capacity: Grok Plus at $8, then Claude Pro at $20. Meta One Core at $7.99 is a cent cheaper but buys creative output, not reasoning.

When two picks tie, run the same three prompts through both and keep the one whose answers you edit least. That test takes ten minutes and beats any ranking, including ours.

Pros and Cons of Each Free Tier

The most common con is the same everywhere: you cannot see your quota before you hit it. That makes the free tiers fine for exploration and risky as the only tool behind a deadline.

Migration Guide: Switching Chatbots Without Losing Your Work

Moving from one free chatbot to another costs nothing but time. The trick is to carry over context, not just chats. Most assistants now keep some form of memory, and our look at how each assistant handles memory explains what transfers and what does not. Follow these steps:

1. List what you would lose. Note your saved instructions, memory entries, projects or custom assistants, and any conversations you still reference.
2. Export your data. Use the data-export option in your account settings if the app offers one, and save the files somewhere you control.
3. Copy your custom instructions. Paste them into a text file. They are short, and they define how the assistant talks to you.
4. Write a context handoff. Ask the old assistant to summarize who you are, what you work on and your preferred style, using the prompt below.
5. Seed the new assistant. Paste the handoff into a fresh chat and ask it to confirm what it understood.
6. Re-run three real prompts. Pick recent tasks and compare answers side by side. Keep the assistant that needs less editing.
7. Check privacy settings. Review training, history and memory toggles on the new account before you share anything sensitive.
8. Delete or lock the old data. Remove chats you no longer need and use lock features where they exist.

Here is a handoff prompt you can paste into the old assistant:

Write a portable profile of how I use you. Include my role, my main projects,
the tone and format I prefer, terms I use often, and mistakes to avoid.
Keep it under 300 words and write it so a different AI assistant can follow it.
Do not include passwords, account numbers or private contact details.

Privacy checks before you switch

Review three settings in every app: whether your chats may be used to improve models, how long history is kept and whether memory is on by default. Ads add a fourth check, since a free account on an ad-supported plan may handle usage signals differently from a paid one. The vendors’ own pages are the source of truth here, and they change often.

Verdict: Which Free AI Chatbot Should You Open First?

Start with ChatGPT Free if you want one tab for everything. It runs a current-generation model in GPT-6 Luna, it leads Pew’s usage numbers by a wide margin (59% of US teens against 23% for Gemini) and its plan ladder is clearly named. Keep Claude Free as your second tab for writing and long documents, and use Gemini when your work already sits in Google’s apps. Those three cover most readers.

Our ranking for a general reader, with the reason for each slot:

1. ChatGPT Free, for breadth and a current Luna model.
2. Claude Free, for writing quality and a fresh 5.5 model family.
3. Gemini Free, for Google integration.
4. Meta AI, for free access inside apps used by billions.
5. Grok Free, for a transparent price ladder.
6. Copilot, for Microsoft 365 users first and everyone else second.

Grok earns a place for tinkerers and budget upgraders. It is the only vendor with every consumer price published, and the free tier includes coding-agent access. Meta AI is the right pick if you create social content, because the free tier lives where your audience does. Copilot belongs to Microsoft 365 users and not yet to anyone else.

The honest limit on all of this is the missing quota data. No vendor publishes a hard number of free messages, so the best test is your own. Run your real tasks for a week and note when each assistant stops you. Pay only for the one that stops you most often, and do it at the cheapest tier that removes the wall.

For more model and chatbot comparisons, browse our AI and machine learning coverage.

Frequently Asked Questions

What is the best free AI chatbot in October 2026?

For most people it is ChatGPT Free, which runs GPT-6 Luna and has the broadest use among the assistants Pew tracked. Claude Free suits writing, Gemini suits Google users, and Meta AI suits social content.

Do free AI chatbots have message limits?

Yes, but none of the vendor pages we reviewed publishes an exact daily count. Grok describes its free tier as limited daily access, and Meta describes allowances as variable. Expect limits to change without notice.

Which free chatbot uses the newest model?

None gives free users its top model. ChatGPT Free reaches GPT-6 Luna but not GPT-6.1 Sol. Gemini 4 Argon and Claude Duet sit on paid plans, and Muse Wonder Full is tied to Muse Code subscribers.

What is the cheapest paid upgrade?

Among verified prices, Meta One Core costs $7.99 a month and Grok Plus costs $8 a month or $80 a year. Claude Pro costs $20 monthly or $200 billed for a year. OpenAI and Google pages did not expose entry prices in our check.

Is Claude free to use?

Yes, Anthropic offers a free Claude plan. The official pages we checked do not confirm its current message limits, model list or reset schedule, so confirm them in the app before you plan around them.

Are Perplexity and DeepSeek still free?

We could not verify their current plan details from official pages this week, so we left both out of the scored set. Check the vendors’ plan pages directly before you rely on any claim about their free tiers.

Can I move my chats between chatbots?

You can move your context, though not always your history. Export your data where the app allows it, copy custom instructions, and use a handoff prompt to seed the new assistant, as shown in the migration guide above.

Nadia Dubois

AI & Innovation Editor

Nadia Dubois is the AI & Innovation Editor at Tech Insider, where she tracks the rapid evolution of artificial intelligence, from foundation models to real-world enterprise deployment. She previously covered AI and startups for La Tribune and contributed to MIT Technology Review’s European coverage. Nadia specializes in generative AI, AI regulation, and the intersection of technology and European industrial policy. She holds a dual degree in Computational Linguistics and Journalism from Sciences Po Paris.

View all articles